Abstract: Arrangements and method for enabling and disabling cache bypass in a computer system with a cache hierarchy. Cache bypass status is identified with respect to at least one cache line. A cache line identified as cache bypass enabled is transferred to one or more higher level caches of the cache hierarchy, whereby a next higher level cache in the cache hierarchy is bypassed, while a cache line identified as cache bypass disabled is transferred to one or more higher level caches of the cache hierarchy, whereby a next higher level cache in the cache hierarchy is not bypassed. Included is an arrangement for selectively enabling or disabling cache bypass with respect to at least one cache line based on historical cache access information.

Abstract: A method of verifying the passage of a data write across a bus is provided including sending the data write from an originator across the bus to a target, counting the number of data entries received at the target with a counter, and transmitting a return echo write from the target across the bus to a return address. The method further includes attaching the counter value to other data associated with the return echo write and polling the return address. The method allows determination of the completion of a data write by comparing the number of data entries included in the data write with the counter value polled from the return address. Alternatively, in a data streaming environment the progress of a data write may be determined by comparing the number of data entries included in the data write at a select point in time with the counter value polled from the return address. Typical data entries which are counted may include, but are not limited to, bytes, words, double words, or similar data quantities.

Abstract: Method for selectively clearing persistent data such as metadata in a storage subsystem. The method includes identifying one or more metadata tracks to clear, receiving a request from a client to clear the identified metadata tracks, passing the request to a storage device driver, and clearing the identified metadata tracks. The metadata tracks to be cleared may be identified by a specific storage function associated with the metadata to be cleared or by a specific storage target associated with metadata. Alternatively, both a storage function and a storage target may be specified. Preferably, the request to clear one or more identified metadata tracks may be executed without affecting unrelated storage functions. In addition, the request may be executed while preserving metadata tracks unrelated to the identified metadata tracks.

Abstract: Methods, systems, computer program products and methods for deploying computing infrastructure for clearing metadata in a storage subsystem. The method includes identifying one or more metadata tracks to clear, receiving a request from a client to clear the identified tracks, quiescing any active processes, clearing the identified tracks and resuming the quiesced processes. The request may be a single request operable to clear all metadata tracks. Alternatively, target volumes having metadata tracks to be cleared are identified and a request is received from a client for each target volume. A dynamic function list may also be created for each target volume, the list including list of functions in which the target volume is associated. After active processes are quiesced, each function in the dynamic function list of each target volume is invoked whereby the metadata tracks are cleared. The quiesced processes are then resumed.

Abstract: Methods, systems, computer program products and methods for deploying computing infrastructure for managing metadata in a storage subsystem are provided. A call is received from a client to update a portion of a metadata track. Data is also received from the client identifying at least one field of the metadata track to be updated, associated update values for the at least one field to be updated and an update type for the at least one field. A data structure field of a client interface is then populated with the identity of the at least one field and the associated update values and an opcode field of the client interface is populated with the update type. A task control block (TCB) is then allocated to the client call and, at the direction of the TCB, various operations are performed without further activity by the client. The metadata track may also be validated before being updated to ensure that the client has access only to valid updates.

Abstract: Methods, systems, computer program products and methods for deploying computing infrastructure for managing metadata in a storage subsystem are provided. A first metadata track is staged from disk storage to a cache storage after which a journal entry is stored in a nonvolatile storage (NVS). The journal entry includes an opcode and update data for the track. The opcode identifies the type of update to be performed and the number of tracks to be updated in the operation. Each of the other metadata tracks is staged and a corresponding journal entry stored. The journaled updates are then applied to the respective metadata track in the cache storage and the tracks destaged from cache to the disk storage.

Abstract: A data processing unit including a first active cooling device configured to respond to a first control signal and a second active cooling device configured to respond to a second control signal. The control signals may be any type of control signal suitable to control the operation of the first and second active cooling devices. The data processing unit also includes a first control function selectively capable of providing the first control signal and/or the second control signal, and a second control function selectively capable of providing the first control signal and/or the second control signal. Logic associated with the first and second control functions is included to determine which of the control functions will provide which control signal at a specific time. Logic and/or switching or isolation apparatus is also included to prevent both the first and second control functions from attempting to provide one of the control signals simultaneously.

Abstract: A response to the continuing trend of ever-increasing processor speeds and attendant increases in memory latencies. Broadly contemplated herein are braids and fibers, high-level programming constructs which facilitate the creation of programs that are partially ordered. These partial orders can be used to respond adaptively to memory latencies. It is shown how these constructs can be effectively supported with simple and inexpensive instruction set and micro-architectural extensions.

Abstract: A method of transmitting information from a primary storage control unit to a secondary storage control unit in an asynchronous data copying system. The method includes building multiple descriptor blocks for transmission from the primary storage control unit, and encrypting multiple payloads according to one of n encryption methods. In addition, the n encryption methods are associated with the primary storage control unit and n decryption keys relating to the n encryption methods are associated with the secondary storage control unit. An index to the n encryption keys is associated with the multiple descriptor blocks. The method further includes encrypting each of the payloads according to one of the n encryption methods and indicating, in the index associated with a selected descriptor block, which decryption key relates to the encryption method used for any payload associated with a select descriptor block.

Abstract: A method, controller, program product and service are provided for more efficiently managing a resource queue. Two or more queues are configured to handle workloads of various sizes. Resource elements are allocated from the top of each queue and returned to the bottom. The size of each queue may be selected to provide sufficient resource elements to handle a system's various levels of workload. As the workload increases and all of the resource elements in the one queue are allocated, new resource elements are allocated from the top of the next queue and returned to the bottom. When the workload decreases, resource elements are no longer allocated from the queues used for higher workloads. Thus, retention of historical data in the queues is enhanced while efficient cache utilization is maintained.

Abstract: Mechanisms, and associated methods, for conducting voice analysis (e.g., speaker ID verification) directly from a compressed domain of a voice signal. Preferably, the feature vector is directly segmented, based on its corresponding physical meaning, from the compressed bit stream.

Abstract: Systems and methods for designing storage area network fabric. Preferably included are an arrangement for collecting user requirements on data flows to be supported by the fabric, an arrangement for grouping the data flows into flow groups according to at least one physical location parameter, an arrangement for designing components of fabric for the flow groups, the components being associated with at least one geographical region, and an arrangement for obtaining fabric by joining the fabric components via interconnection fabric, whereby flow groups over a plurality of geographical regions are supported.

Abstract: In a computer network environment, a server protocol is provided to process read requests from clients. Rather than all read requests being processed synchronously or all read requests being processed asynchronously, an attempt is first made to perform a synchronous read. If the synchronous read is unsuccessful, the connection through which the request was received by the server is registered with a monitoring service. When the data is ready to be read, an appropriate callback is called and the data transmitted. An optional delay may be imposed before the synchronous read is attempted to increase the likelihood that the attempt will be successful. A series of delays/read attempts may also be employed in order to increase the likelihood still further that an attempt will be successful.

Abstract: Methods and arrangements for constructing a network overlay via embedding an interconnection network in a network overlay graph. Presented herein is one example of an approximation algorithm for efficient mapping of hypercube topology on a network overlay of N nodes such that the average network delay distance measured on the network overlay is minimized. The network delay distances between pairs of overlay nodes are measured on the shortest network paths. The mapping of the interconnection topology to the network overlay nodes preferably involves assigning 2m m-length binary indexes to the network overlay nodes. There is proposed herein a polynomial time algorithm for interconnection topology embedding using iterative cluster division and node binary index assignment. There is further proposed herein a method for construction of network overlays through interconnection network topology embedding, with the objective of optimizing a quality of service metric (e.g.

Abstract: The state or condition of a data storage drive, or a subsystem within a drive, may be evaluated by comparing a set of selected parameter values, converted into a trial vector, with a number of model or exemplar vectors, each of which was represents a particular state or condition of a sample drive. Examples of such conditions may include “good”, “marginal”, “unacceptable”, “worn”, “defective”, or other general or specific conditions. Sets of parameter values from the drive are converted into input vectors. Unprocessed vectors are then processed against the input vectors in an artificial neural network to generate the exemplar vectors. The exemplar vectors are stored in a memory of an operational drive. During operation of the drive, the trial vector is compared with the exemplar vectors. The exemplar vector which is closest to the trial vector represents a state which most closely represents the current state of the drive.

Abstract: In a power controller or other computing resource shared by multiple processors, an ID is written to the lock register, thereby designating a master processor. A timer is then initialized to count for a predetermined period. Periodically, the master processor transmits a “heartbeat” signal to the shared resource, indicating that its operation remains normal. Upon receipt of the heartbeat signal, the timer is reset and begins a new count for the predetermined period. If the timer reaches the end of the period without having received a heartbeat signal, indicating that the master processor has failed or hung, the lock register is cleared again and an interrupt signal is broadcast to all of the processors. The processors compete for master status anew and, when an ID is successfully written to the lock register, the timer is restarted and the new master periodically transmits its heartbeat signal.

Abstract: The state or condition of a system may be evaluated by comparing a set of selected parameter values, converted into a trial vector, with a number of model or exemplar vectors, each of which was represents a particular state or condition of a sample system. Examples of such conditions may include “good”, “marginal”, “unacceptable”, “worn”, “defective”, or other general or specific conditions. Sets of parameter values from the system are converted into input vectors. Unprocessed vectors are then processed against the input vectors in an artificial neural network to generate the exemplar vectors. The exemplar vectors are stored in a memory of an operational system. During operation of the system, the trial vector is compared with the exemplar vectors. The exemplar vector which is closest to the trial vector represents a state which most closely represents the current state of the system.

Abstract: A method, apparatus, and article of manufacture containing instructions for the management of data in a point-in-time logical copy relationship between a source and multiple target storage devices. The method consists of establishing first and second point-in-time logical copy relationships between a source storage device and at least two target storage devices concerning an extent of data. Upon establishment of the point-in-time copy relationships, a first cache scan request is received relating to the first point-in-time logical copy relationship to remove a first extent of data from cache; a similar cache scan request is received related to the second point-in-time logical copy relationship. The first cache scan request is processed, and the successful completion of both the first cache scan request and the second cache scan request is returned to the storage controller upon the processing of only the first cache scan request.

Abstract: A distributed computer system includes links and routing devices coupled between the links and routing frames between the links. Each of the routing devices includes a congestion control mechanism for detecting congestion at the routing device and responding to detected congestion by gradually reducing an injection rate of frames routed from the routing device.

Abstract: An inline electron holograph method for observing a specimen with a transmission electron microscope having an electron gun, a collimating lens system, two spaced objective lenses, a biprism, and an imaging means comprises the steps of: with the first objective lens forming a virtual image of a portion of the specimen; with the second objective lens focussing the virtual image at an intermediate image plane to form an intermediate image; and projecting the intermediate image onto the imaging means.